- Home
- Skills
- Jpicklyk
- Task Orchestrator
- Manage Schemas
manage-schemas_skill
- Kotlin
166
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
3 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill jpicklyk/task-orchestrator --skill manage-schemas- SKILL.md5.7 KB
Overview
This skill manages note schemas stored in .taskorchestrator/config.yaml for the MCP Task Orchestrator. It lets you create, view, edit, delete, and validate the templates that define what notes agents must fill at each workflow phase. Use it to ensure consistent, gated workflows and to keep schema configuration correct and up to date.
How this skill works
The skill reads or bootstraps .taskorchestrator/config.yaml, parses the note_schemas section, and routes your request to the appropriate operation: CREATE, VIEW/VALIDATE, EDIT, or DELETE. For write operations it performs interactive prompts to gather changes, merges YAML updates, writes the file, and reports the exact changes. For validation it runs structural and semantic checks and returns actionable findings.
When to use it
- Create a new schema to standardize notes for a workflow stage
- List or inspect existing schemas and note definitions
- Edit an existing schema (add/remove notes, toggle required, update guidance)
- Delete a schema you no longer use (with orphan warnings)
- Validate the config after manual edits to find syntax or semantic errors
Best practices
- Back up .taskorchestrator/config.yaml before making manual edits
- Prefer the interactive CREATE/EDIT flows to avoid YAML key duplication or formatting errors
- When you change schemas, reconnect the MCP server (/mcp) so the server reloads updated schemas
- Give notes clear keys, roles, and guidance so agents can fill them reliably
- Keep required notes to the minimum necessary to avoid blocking progress unnecessarily
Example use cases
- "Create schema": walk through an interactive flow to define phases and notes for a new feature pipeline
- "Show schemas": display a summary table of all schemas with per-phase counts and required-note counts
- "Add a note to schema": append a required test-plan note to the bug-fix schema with description and guidance
- "Delete schema": remove an obsolete schema and warn about items that may have orphaned notes
- "Validate config": lint a hand-edited config.yaml and get a list of fixes and suggestions
FAQ
For create/edit/delete operations the skill bootstraps the directory and a config.yaml with note_schemas: present. For view/validate it reports that no schemas are configured.
Why do schema changes not take effect immediately?
The MCP server caches schemas on first access. Run /mcp to reconnect the server so it reloads the updated config.